This is a second revision of a collaborative R01 four-year competing continuation proposal to create a large repository-based sample of cases with recurrent, early-onset major depressive disorder (MDD-RE), and to use positional cloning to identify depression susceptibility genes in regions of significant linkage in our genome scan. The completed four-year project collected 680 families containing 927 affected sibling pairs (ASPs) (MDD-RE diagnostic model) and additional affected relatives (GenRED I). Blinded clinical data and blood specimens for cell culture were deposited in the NIMH repository and are being made public. Linkage fine-mapping has demonstrated genome-wide significant linkage on chromosome 15q; in the 10 cM genome scan, suggestive sex-specific linkage was observed in three regions (6p-q, 8p, 17p), with the result on chromosome 17p approaching genome-wide significance. Six collaborating sites now propose to: (1) Collect (during Years 1-3) an additional 1,350 European-ancestry (EUR) MDD-RE probands (GenRED II) meeting identical criteria (including evidence of having an affected sibling) to create a total repository sample of 2,000 EUR MDD-RE cases, plus cell lines/DMA from available parents, unaffected sibs and male-male ASPs. (2) Initiate a repository-based collection of African-American (AA) MDD-RE probands meeting the same clinical criteria. We will collect 750 AA probands plus available parents and affected siblings, with involvement of young minority co-investigators; AA controls will be available from the repository. A site at Howard University has been added to lead this effort. AA recruitment will continue through Year 4 to build the repository sample. (3) Collect data on childhood abuse and neglect and parental loss, major environmental MOD risk factors; (4) Carry out linkage fine-mapping studies of chromosomes 17p, 1q, 5q, 6p-q, 8p and 18q to maximize evidence for linkage and to narrow candidate regions. (5) Carry out linkage disequilibrium (LD) mapping and intensive gene analysis studies in the 15q candidate region and one additional region in 2,000 EUR cases and 2,000 screened, ethnically-matched controls; and carry out LD fine-mapping studies in the most significant genes in 600 AA cases (the N available early in Year 4) and 1,000 controls, using high-throughput SNP genotyping methods, to identify a depression susceptibility gene. The proposed studies will contribute to the understanding of this devastating common disorder by identifying susceptibility genes, and by creating a public collection of biological materials and clinical data, as well as over 13 million SNP genotypes, to facilitate further investigation of recurrent MOD and related phenotypes.